1. Field of the Invention
The present invention relates to a vector which is integrated into the chromosome of yeast, specifically Candida utilis, with a high number of copies and which can be stably maintained even under nonselective culture conditions. The present invention also relates to heterologous gene expression, specifically the high-level expression of a single chain monellin and amylase, using said vector, and further relates to a method for isolating and purifying a single-chain monellin from single-chain monellin-producing recombinant yeast cells.
2. Background Art
In order to prepare gene products in large quantities using recombinant DNA procedures, it is necessary not only to select an appropriate host but also to increase the number of gene transcripts, to improve the translation efficiency, and to improve the stability of proteins produced in the various steps of gene expression. In order to increase the number of gene transcript for high-level production of gene products, it is necessary to use a highly effective transcription promoter and it is also important to increase the number of copies of the gene-expression unit which consists of the transcription promoter/terminator sequence and the gene to be expressed, thereby increasing the number of transcripts as a whole. Furthermore, for industrial scale production, it is extremely important that the gene-expression unit be stably maintained in the microbial cells. Plasmid vectors are at a disadvantage in this regard and generally stabilized by integration into a chromosome.
Regarding several yeast species other than C. utilis, it has been recently reported that dozens of copies of a vector could be integrated into the ribosomal RNA gene (rDNA) regions by using the vector carrying a transformation marker gene in which the promoter region was trancated to reduce the expression level (Lopes T. S. et al., Gene, 79, 199-206, 1989; Bergkamp R. J. M. et al., Curr. Genet., 21, 365-370, 1992; Le Dall M. T. et al., Curr. Genet., 26, 38-44, 1994).
However, it has been shown that to achieve high-copy-number integration into the chromosome, it is necessary to integrate the vector into the ribosomal RNA gene regions; otherwise, a large number of copies will not be obtained when the vector is integrated into other gene loci (Lopes T. S. etal., Gene, 105, 83-90, 1991). It has further been reported that the introduced genes might not be sustained due to recombination between their repetitive sequences because the integrated vectors existed in a tandem form in the chromosome (Lopes T. S. et al., Yeast, 12, 467-477, 1996). In particular, when microbial cells are cultured under nonselective conditions or microbial growth is slow (for example, when the expression product is present in abundance in the microbial cells), successive cultivation for generations will result in an increase in the ratio of cells without vectors. Accordingly, when recombinant yeasts are cultured under nonselective conditions (particularly in a large-scale culture), stable maintenance of the integrated vectors is of extreme importance. It has been reported that an expression unit integrated into the chromosome was stabilized by shortening the size of vector DNA (Lopes T. S. et al., Yeast, 12, 467-477, 1996).
Candida utilis, a yeast which efficiently assimilates pentoses such as xylose, has been approved to use as a food additive by the Food and Drug Administration (FDA) along with Saccharomyces cerevisiae and S. fragilis. A transformation system for Candida utilis using homologous recombination was developed recently, and heterologous protein production was reported (WO/95/32289). However, further improvements as to high-copy-number introduction of the vector into the chromosome and stabilization of the expression units are still to be achieved.
Sweet proteins are expected to be extensively used as a highly safe, low calorie sweetener, food additive, or sweetening agent in foods, drugs and the like, and even animal feed. Examples of such sweet proteins include monellin and thaumatin.
Thaumatin is a protein which can increase the palatability of food (i.e., enhance flavor and aroma) and is extracted from seed coats of the fruit of plant, Thaumatococcus daniellii Benth. However, although it is commercially available, the industrial use of plant-derived thaumatin is extremely limited because of the scarce availability of fruit for extraction. Although the production of thaumatin in a number of microbial hosts has been tried to date, published reports would indicate that expression of the protein was extremely difficult, and the protein so obtained was of minimal sweetness (Zemanek E. C. and Wasserman B. P., Critical Reviews in Food Science and Nutrition, 35, 455-466, 1995).
Monellin, a protein found in the fruit of the tropical plant Dioscoreophyllum cumminsii, is more than 2,000 times sweeter than sucrose on a weight basis, and its amino acid sequence is known. This protein comprises two nonhomologous subunits, A and B, and its tertiary structure has been reported (Hudson G. et al., Biochem. Biophys. Res. Comm., 71, 212-220,1976; Ogata C. et al., Nature, 328, 739-742, 1987; van der Wel H., FEBS Letters, 21, 88-90, 1972; Morris J. A. et al., Biochim. Biophys. Acta. 261, 114-122, 1972; Bohak Z. et al., Biochim. Biophys. Acta., 427, 153-170, 1976; Frank G. Hoppe-Seyler's Z. Physiol. Chem., 357, 585-592, 1976). Natural monellin rapidly loses its sweetness at high temperatures at acidic pHs. Attempts are under way to produce a more thermally stable protein which retains its sweetness by linking the two chains comprising monellin, namely, linking the N-terminal of the subunit A with the C-terminal of the subunit B to make a single polypeptide chain (Japanese Patent 1990/504028; Japanese Patent Laid-open 1993/70494; Kim S-H. etal., ProteinEngineering, 2, 571-575, 1989). This single-chain monellin having excellent properties is being expected for use in food as a low calorie, highly stable protein sweetener, a food additive in place of conventional sweetening agents, or a sweetening agent.
However, as long as the present inventors know, the large-scale microbial production of monellin has not been reported.